On Sat, Dec 12, 2020 at 08:24:24AM +0100, Ard Biesheuvel wrote:
> On Sat, 12 Dec 2020 at 07:43, Eric Biggers wrote:
> >
> > Hi Ard,
> >
> > On Tue, Nov 03, 2020 at 05:28:09PM +0100, Ard Biesheuvel wrote:
> > > @@ -42,24 +42,24 @@ static void chacha_doneon(u32 *state, u8 *dst, const
> > > u8 *src,
On Sat, 12 Dec 2020 at 07:43, Eric Biggers wrote:
>
> Hi Ard,
>
> On Tue, Nov 03, 2020 at 05:28:09PM +0100, Ard Biesheuvel wrote:
> > @@ -42,24 +42,24 @@ static void chacha_doneon(u32 *state, u8 *dst, const u8
> > *src,
> > {
> > u8 buf[CHACHA_BLOCK_SIZE];
> >
> > - while (bytes >= CHA
Hi Ard,
On Tue, Nov 03, 2020 at 05:28:09PM +0100, Ard Biesheuvel wrote:
> @@ -42,24 +42,24 @@ static void chacha_doneon(u32 *state, u8 *dst, const u8
> *src,
> {
> u8 buf[CHACHA_BLOCK_SIZE];
>
> - while (bytes >= CHACHA_BLOCK_SIZE * 4) {
> - chacha_4block_xor_neon(state,
On Tue, Nov 03, 2020 at 05:28:09PM +0100, Ard Biesheuvel wrote:
> The current NEON based ChaCha implementation for ARM is optimized for
> multiples of 4x the ChaCha block size (64 bytes). This makes sense for
> block encryption, but given that ChaCha is also often used in the
> context of networkin
The current NEON based ChaCha implementation for ARM is optimized for
multiples of 4x the ChaCha block size (64 bytes). This makes sense for
block encryption, but given that ChaCha is also often used in the
context of networking, it makes sense to consider arbitrary length
inputs as well.
For exam
